Tarceva phase III data in NSCLC was presented at ASCO in 2004. The trial was Tarceva vs placebo in 2nd and 3rd line NSC lung cancer. 50% were adenocarcinoma, most of the rest were squamous. PFS was 2.2 months in the Tarceva arm vs 1.8 months in the placebo arm. That is equivalent to about 10 wks vs 8 wks, about a 25% improvement vs placebo. That result was statistically significant because the trial contained 731 patients. On the day those results were initially released, OSIP's market cap exceeded five billion dollars, based almost entirely on prospects for Tarceva. Tarceva was later approved on the basis of that data.
Now consider ARQ 197 data in NSCLC. Although a much smaller trial (167 pts), it was also a randomized, double blinded placebo controlled trial, with about 30% squamous histology patients. PFS was 16.1 weeks vs 9.7 wks, representing a 65% increase in median PFS in the ARQ 197 group vs Tarceva monotherapy. In the non-squamous group, about 117 pts, there was a 100% improvement in PFS in the ARQ 197 group. If replicated in a larger trial those numbers seem almost certain to be statistically significant.
Note that the Tarceva PFS of 9.7 weeks in this trial was similar to the 10 weeks PFS in the phase III Tarceva trial from 2004, suggesting that the populations were fairly similar and Tarceva performed about as expected. This seems to indicate that c-Met inhibition had a substantial additive effect with Tarceva in NSCLC, possibly an even greater effect on PFS than the approved drug itself.
It would have been nice to get some response rates from today's press release, and the absence raises the possibility that ORR was not improved in the ARQ 197 group. But, in early trials the response rates from another anti-angiogenic drug, Avastin, were usually not impressive either. It was the PFS and OS data that led to Avastin's approval.
Yes, these are phase II data and they're early. But, as phase II trials go, this one was a randomized, controlled, blinded trial, which gives it more weight than a lot of the single arm trials that end up not being validated. Plus, the treatment effect was achieved vs an active control arm in a patient population that is known to be sensitive to Tarceva.
All in all, at its current 250 million dollar market cap, shares of ARQL are basically a call option on future success of ARQ 197 in NSCLC and potentially other large cancer indications. The company has plenty of cash, not much debt and a partner to split development costs, which means they can avoid massive dilution. This stock looks cheap to me. I took a small position late today and will add gradually if it goes down.
Tarceva is contraindicated for squamous and mutated K-ras.
I like ArQule's path toward a tightly targeted phase III trial. Targeting chemo now seems to be at the top of ASCO's list of ways to improve cancer therapy in the future. ArQule may become one of ASCO's 2010 highlights on how to design future phase II trials.
10-50% because we dont know the p values. Did they just miss the stat significance (e.g., is p = 0.06) or is it >0.2? My first reaction was p must be <0.1 because adjusting for imbalances in the distribution of key prognostic factors caused p to become stat significant. However, doing so causes a significant change in the HR. So maybe not.
Actually, I'm more interested in one piece of the puzzle that they did not publish.
What is the HR for the 30% squamous histology patients? I prefer if it were close to 1 (i.e., the drug doesnt work for this subgroup), which would explain why they missed stat significance. If that's the case the p value of the non-squamous patients might be very (very) low. This increases the chances of P3 being a success. I would rather invest in something that has a better chance of success than a larger market size.
No matter how good the P2 data is, I learned not to give more than 50% chance to a P3 trial.
I agree what you say. ARQ-197 is quite different from others and the potential is enormous.
But we are investors and biotech seldom follows logic. So I still think it will drop below $5 for me to pick some more. However, I already bought about 50% of my original investment.
If they missed the asco deadline, the next opportunity is august and I think I'll have my $5 by then.
Are you in your max or waiting for it to drop as well?
The phase II results will lead to allow much better targeting of patients most likely to benefit.
K-ras results were called the surprise of the stury. "Surprise":
* ARQ-197 potency in wild type proved stronger than expected
* ARQ-197 actually showed potency in mutated K-ras? Keep an eye on Pancratic Cancer trial, mutated K-ras playes in a big role in the most difficult to treat cancer.
I suspect it's good news as they are anxious to present the K-ras data at a professional conference in the future. ASCO?
Any thoughts good or bad?
>But unfortunately, we’ve seen this before, and we’ll see it happen again.
Before seeing full results (notice that they never told us the p values), I give 10-50% chance that the results of this trial being reproduced in a P3 trial.
However, even that makes arql an excellent long term investment.
1. the non-squamous group, ~117 pts (which is quite large), HR is OK and is stat significant
2. after adjusting for imbalances in the distribution of key prognostic factors, HR is OK and stat significant
3. They never told us about the HR of non-squamous group after adjusting for imbalances (i.e., 1 + 2), which should also be OK and stat significant (but you never know).
These are better than some of my other biotech investments.
Disclaimer: Yesterday, I sold all and bought some back. I'm waiting for it to drop before I buy more (I believe it will ease back in the next one month). Also, I'm praying that sarcoma wont be released before I buy more or comsi comsa sarcoma data if it does.
I thought of shorting yesterday but I did not want to risk it as the management may releases sarcoma data earlier than expected.
single arm & multiple countries
ASA404 had two warning signs about phase II trials.
Same with Medivation and its Alzheimer drug called Dimebon. Phase II data was centered in Russia with who knows what kind of oversight.
ArQule took the slow and careful route double blinding, crossover....
After a couple of major blowups in March I can see why actual investors (including institutions) could have been spooked and been glad to book a double. I was impressed with the design of ArQuele's phase II. Remembering it acually it's hazard ratio ((failed)). Sell first and ask questions later ruled the day.
Thanks for all information. It would take another 4-5 years to find out whether or not ARQ197 would work in this sitting. I hope people can think more logically and learn from this news:
Unfortunately, yesterday brought bad news in the form of another negative phase III trial for advanced NSCLC, this one the ATTRACT-1 trial of carboplatin/taxol (paclitaxel) with or without ASA404, which is in a class of drugs called “vascular disrupting agents” that I described in a post back in 2007. There had been a phase II trial of this combination with ASA404 that looked very promising, with a median overall survival of 14 months for the combination arm, vs. only 9 months for the arm getting chemo alone. This led to a couple of larger, more definitive studies that have been ongoing. The first, ATTRACT-1, was in the first line setting, and as noted above gave ASA404 on a backbone of carbo/taxol, and has completed enrollment. The second, ATTRACT-2, is still accruing patients and is giving taxotere (docetaxel) with or without ASA404 in the second line setting.
The news yesterday was that the ATTRACT-1 trial is being halted (no more treatment with the study drug, and people are no longer required to adhere to treatment according to the protocol-based plan), because an interim analysis of the trial results showed that the investigational arm wasn’t doing any better than the standard chemo arm, and it appeared that there was no way for this to improve to the point at which ASA404 would confer a survival benefit. We don’t have more details available than that, though this was enough information to lead shares of the UK-based company Antisoma, who make ASA-404, to lose 2/3 of its value over the day. Novartis had been co-developing the drug with Antisoma.
My understanding is that at the present time the second line trial is ongoing and is planned to continue to accrue patients, though I have to imagine that the enthusiasm for the concept is much lower right now that the first line trial is being abandoned. Moreover, when the CEO of Antisome says, “We have to say it’s quite unlikely that we will get any future cash flows from this programme“, that doesn’t exactly motivate people to enroll on the sister trial.
We certainly need to learn more about the trial and whether there may be subsets of patients who did better with ASA404, but if there is something to learn from the experience, I think I would highlight that this was an example of a very glowing press release about supposed breakthrough results from a phase II trial. Here’s what I said in my post from 2007:
They call it a “positive trial”, which is really a wording I’d reserve for a larger phase III trial that can actually detect differences between one treatment and another, and I would consider it rather aggressive to make an international declaration of the astounding benefit of their drug based on a phase II trial. Many new drugs and combinations show phenomenal results in small phase II trials reported early, and many don’t pan out. This doesn’t mean I don’t think the results are encouraging — I absolutely do, and I’d consider this agent on my “ones to watch” list for years to come.
I wish ASA404 had been the exception and not the rule, because we could certainly use more effective agents for lung cancer. I would have been very happy to have had my skepticism been misguided. But unfortunately, we’ve seen this before, and we’ll see it happen again. I’d encourage people to be hopeful, because all of our positive trials were preceded by encouraging early work in smaller studies, but unfortunately many breathless press releases turn out to be more sizzle than steak.
ASA404 is the NSCLC trial that failed phase III
Phase II trial, single arm & multiple countries:
This was a single-arm trial that enrolled patients receiving first-line chemotherapy treatment for stage IIIb or IV NSCLC. Thirty patients received up to 6 cycles of standard therapy (carboplatin AUC 6 mg/mL*min and paclitaxel 175 mg/m²) plus ASA404 1800 mg/m². The trial was conducted at hospitals in New Zealand, Germany and Australia that had also participated in a previous randomised, controlled study comparing standard therapy plus ASA404 1200 mg/m²with standard therapy alone
MiT tumors, which include clear cell sarcoma (CCS), alveolar soft part sarcoma (ASPS) and translocation-associated renal cell carcinoma (RCC), are linked biologically through a common chromosomal abnormality that is responsible for the over-expression of c-Met resulting in the development of these tumors. Tumors with this abnormality are resistant to current therapies and, in the absence of successful surgical resection, are invariably fatal.